Epidemiological studies suggest that repetitive loading is of importance in the development of back injuries including, disc injuries. Yet, the present information from the epidemiological literature does not allow dose-response relationships to be developed, thereby making it challenging to develop accurate guidelines for prevention of occupational musculoskeletal disorders of the spine. Clearly, in vitro studies are limited when it comes to furthering our understanding of the complex relationships that exist between the large number of parameters needed to describe the effect of single load application and how disc tissues break down over the course of many years of exposure. We believe that an improved understanding of these complex relationships can best be achieved by numerical techniques such as the Finite-Element Model (FEM). Using the refined FEM we will test the following hypotheses: (1) The disc is most resilient to the effects of repetitive loading when the loading mode is compression only. This would be analogous to a repetitive lifting task wherein the torso is maintained in an upright non-twisted posture. (2) Damage will accumulate more rapidly when the cyclic loads create spinal segment motions that include combinations of forward bending, twisting, and lateral bending as would be found during most asymmetric lifting tasks as compared with cyclic loads in which there is only forward bending (symmetric lifting). (3) The damage accumulation process is dependent upon the amount of variation in the magnitude of the lifted load. (4) The damage accumulation process is dependent upon the variability in the inter-lift interval. These hypotheses will be addressed with the help of the following specific aims : 1. We will refine and validate using cadaver studies an existing lumbar spine FE model by including fluid flow among various motion segment components. 2. We will add muscle force vectors to the validated dynamic FEM refined in the first aim so that actual load history data obtained from dynamic lifting tasks can be used to drive the loading of the model. 3. We will use the dynamic muscle vector FEM to quantify damage accumulation patterns in the lumbar disc due to cyclic loads with different types of load histories (Hypotheses 1 and 2). 4. We will extend the muscle vector FEM to predict damage accumulation from repetitive loading when there are stochastically modeled variations in the load parameters similar to those found in repetitive material handling jobs. (Hypotheses 3 and 4).